Blindness test flawed?

By Fiona Douglas

As a breeder of miniature dachshunds, a test that has met my scrutiny is that for Progressive Retinal Atrophy (PRA) which is a disease that causes blindness. I believe that breeders of smooth haired dachshunds could be suffering unnecessarily if the DNA disease screening test proves to be flawed.

Responsible breeders take every opportunity to ensure the health status of their breeding dogs. Indeed, in Victoria it recently became law that dogs with heritable diseases can only be bred under an approved breeding program; a good thing providing, that is, the disease tests are correct.

Costs involved with incorrect diagnoses are many. One of the most serious of these is decreasing the health of a breed. Purebreds already have relatively small gene pools. Erroneous tests make this problem worse by seeing the phasing out of genetics that were in fact sound. The dangers of a smaller gene pool are the increased risks of new genetic defects arising because long-buried recessive genes surface into the phenotype.

Costs to the breeder include the personal distress of being told that a fine dog of wonderful temperament, who is perfectly fit and healthy, is to be publicly tagged ‘breed with conditions’. This naturally impacts breeding programs: programs that might have involved many, many years of planning and dedication. Then there is the financial cost of finding new breeding stock (thousands of dollars), not to mention the high cost of the DNA test itself which can be around $80 for an uncertified test result or up to $160 for a result certified by a vet. Multiply that by 10 dogs, say, and the costs build up.

Why do I doubt the PRA test?

PRA is cited as a problem disease in some breeds and/or types within breeds. That is, the disease presents with greater frequency in these particular breeds or types when compared with the global purebred dog population.

In the miniature dachshund, it has long been accepted (although now, I do wonder) that the long haired mini can develop this condition. Accordingly, when a DNA test for PRA became available, responsible breeders took the opportunity to have their dogs tested. I tested my long haired dogs and found three to be ‘clear’ and one to be a ‘carrier’. A fortunate outcome, any breeder would agree.

At the time I was having these tests done, I was approached (via my vet) to be part of a trial for a new DNA test for PRA in smooth haired dachshunds. I was a bit surprised, because I was unaware that PRA was a problem in smooths. I went along with the trial though, assuming my knowledge was incomplete in this regard. Three of my mini smooths were tested for PRA (the test sample is a cheek cell swab, so no big deal for the dogs).

When the results came back, I was staggered to see all three smooth minis tested as ‘carriers’. But how could this be? My dogs came from all over Australia and had no common genetics to best knowledge. My vet concurred, and we concluded that the DNA test was flawed and it would be ‘back to the drawing board’ with that one.

In the interests of an accurate public record, I approached the diagnostic company that had performed the trials to ensure that the names of my smooths were not on their records as ‘carriers’, for clearly the test was invalid. This request was declined by the company, with my being told that the test was, in fact, valid.

The company’s validation was unscientific, to say the least. Namely, regarding my request to have my dogs names removed from ‘carrier’ status, their Manager of Animal Diagnostics replied:

“Although the test works on miniature longhaired dachshunds I see no problem with it being relevant to smoothhaired. We test and identify the mutation on the report. I did perform the tests for free as part of attempting to see whether the test was valid for smooths and we identified them as carriers, which lets me believe that the test is valid for that breed. If it did not work or was not valid for that breed it would have come back as CLEAR.” [sic.]

In other words, it seems the test was merely shifted across to another breed type based on an ‘assumption’ that appears to me to be lacking in science or logic. Why should this test be relevant to smooth haireds? What research was done to see if smooth haired dogs even display the disease?

Bunkum – figures don’t add up

So improbable did this seem to me that it set me on a hunt to unearth some basics. Like how many blind smooth mini dachshunds are actually out there? Or, in scientific language, what is the incidence of this disease in the mini smooth dachshund population; what is the epidemiology of PRA in mini smooth dachshunds? If I find out this basic information then it’s a straightforward matter to work backwards to see if the DNA test is valid.

This is how my logic went.

Firstly, identify the proportion of mini smooth dachshunds that supposedly have PRA blindness according to the DNA test results conducted to date.

Next, investigate the actual proportion of the mini smooths in the real world displaying PRA blindness.

If the DNA test is valid, these two figures should be near identical. That is, the proportion of dogs afflicted by blindness will be not significantly different to the proportion indicated by the DNA test results.

Putting it simply, we are going to check out if the theoretical world matches the real world. And if it doesn’t, then the test is (in my opinion) worse than useless.

Extent of blindness according to the test

The DNA test companies explain that the PRA disease follows simple inheritance (that is, Mendelian dominant/recessive inheritance laws) and that the defect gene is recessive. The DNA test is for this defective gene and a dog can get one of three results:

• Clear – no recessive gene, so cannot get the disease nor pass it on to progeny,

• Carrier – one recessive gene, so cannot get the disease but can potentially have diseased progeny;

• Affected – has two recessive genes and therefore has the disease and can potentially pass this on to progeny.

Given a big enough sample of DNA PRA test results, we can easily work out the proportion of the smooth dachshund population that should currently be blind or, if not blind, sight impaired to the extent that their condition warrants being labelled a disease or defect. Afterall, that’s what the test is supposed to be determining: the existence of the PRA blindness in dogs and the chances of PRA blind progeny[1] arising from their matings.

Fortunately we find we have a big enough test sample to call upon. Namely, the UK Kennel Club has recorded a widespread screening of miniature smooth dachshunds involving about 1000 dogs, whose test results have been published on the web alongside their registered names and those of their dams and sires.

The results are:

• 36% of smooth mini dachshunds tested ‘Clear’;

• 46% of smooth mini dachshunds tested ‘Carrier’; and

• 18% of smooth mini dachshunds tested ‘Affected’.

This tells us that about 18% of miniature smooth dachshunds should currently be afflicted with PRA to the extent that it warrants being labelled a disease or heritable defect. That is, 18% of mini smooth dachshunds should display diminished quality of life due to vision loss.

Extent of blindness in the real world

Armed with the amazingly high figure that nearly 20% of all mini smooth dachshunds should have blindness to some debilitating degree, the task of finding figures for dogs displaying this disease should have been a simple one.

But try as I might, I couldn’t find information or figures about real-life blind smooth haired mini dachshunds. Plus the DNA test company blurbs certainly don’t cite the occurrence of the disease in the real world; not that I could find at any rate.

The closest I came to finding evidence of PRA affected smooth haired dachshunds (mini or otherwise) were irrational articles saying such things as ‘PRA blindness in the mini smooth dachshund develops at such a great age that they have probably already died from something else’. Good grief: so PRA is considered a problem to a deceased dog? Or other equally irrational articles saying that the signs of the disease in the dog are ‘so subtle that the owners can’t even tell’. Good grief again: if it’s so subtle, how can it be a problem?

What I did come across was anecdotal evidence supporting the irrationality of these last two points:  owners of mini smooths that tested as ‘affected’ have said that their dogs have no signs of having the disease.

Clearly a fuller survey of the 178 (18%) ‘affected’ dachshunds in the UK test pool of 1000 is in order, for it would be the true decider here. How are they travelling? Are they suffering loss of sight? Have the DNA testing companies profiting from the PRA test to mini smooth dachshunds conducted such a survey? If no, why not?

Frustrated, I went back to my vet again and he reiterated that PRA is not a problem in smooth haired mini dachshunds.

Next I approached one of Australia’s leading canine eye specialists, Dr Robin Stanley[2]. Our brief phone conversation (of 8/2/2010) went like this:

Fiona: Does the mini smooth dachshund get PRA?

Dr Stanley: No. The mini smooth dachshund is not known for PRA.

Fiona: So even though there is a test for the mini smooth that has become available, that doesn’t mean they get the disease?”

Dr Stanley: Correct: just because there is a DNA test for the smooths doesn’t mean that the disease is prevalent.

Science fiction vs real world

Breeders and dog associations are actively trying to do the right thing and breed healthy dogs. They are under the watchful eye of the public, and rightly so. But if tests are not valid and don’t hold up to the most elementary scientific scrutiny, then the tests are potentially worse than useless: they are causing harm.

In summary, it is my assessment that the DNA test results for PRA in miniature smooth haired dachshund (as evidenced in the UK kennel Club sample set) indicates that 18% of this breed type should currently be displaying this disease. Yet this statistic is incorrect because 18% of mini smooth dachshunds do not suffer sight loss, be that mild, severe or otherwise. Accordingly, I believe the test to be invalid.

I stand happy to be corrected and welcome any comments.

Post Script
As breeders we need to question science-based providers with simple commonsense questions. If the facts don’t add up, then it is up to the test provider to explain the logic clearly, for so very much is at stake.

Footnotes

[1] Interestingly, if you consider an F1 from this sample population of dachshunds (assuming any PRA status dog has equal chances of mating with any other PRA status dog among the selected sample set) then it suggests that PRA in mini smooth dachshunds is escalating at an exponential rate, with the incidence of the disease more than doubling in the next generation from 18% to 38%.

[2] Dr Robin Stanley – Ophthalmologist. Dr Stanley is an honours graduate of the University of Melbourne in 1984. He is a Member (small animal surgery 1988) and Fellow (ophthalmology 1990) of the Australian College of Veterinary Scientists. He is a clinical Consultant & visiting Senior Lecturer at the School of Veterinary Science, University of Melbourne and a tutor for the Centre for Veterinary Education, University of Sydney. He has lectured throughout Australia, and also in New Zealand, Japan, Singapore, Taiwan, Thailand, Malaysia, Hong Kong, Sweden, England, South Africa and the United States.